Because of the limited laboratory facilities in developing countries, HIV plasma RNA levels are not used to monitor therapy in these settings. Rather, clinical signs and symptoms and CD4+ T-cell counts are used to monitor response to therapy, which results in a delay in the recognition of viral replication and detectable HIV RNA levels in patients failing therapy. We hypothesize that, during this delay, patients have an increase in the absolute number of HIV-1 drug resistance mutations, develop HIV-1 resistance to a greater number of antiretroviral drugs, and incur significant morbidity and mortality. Further, this delay in recognition of virologic failure may create a reservoir of HIV-1 infected patients at risk for transmitting drug resistant virus to others in the general population. The aims of the proposed research are: Specific Aim 1: To compare HIV-1 drug resistance and clinical status at the time of detectable HIV plasma RNA (virologic failure) and at the time of WHO-defined clinical/immunologic failure in patients receiving antiretroviral therapy in Haiti. We will enroll and follow a cohort of 200 HIV-1 infected adults from their initiation on antiretroviral therapy at the GHESKIO Center in Port-au-Prince, Haiti. Patients will receive care according to WHO guidelines. Viral loads are not available in "real time" for monitoring therapy in Haiti. Laboratory monitoring is done by CD4 T cell count every 6 months. A plasma sample will also be frozen for research purposes every six months. We estimate that 50 patients will fail therapy by clinical and CD4 T cell criteria between 6 months and 48 months of follow up. We will perform HIV-1 RNA levels on all frozen plasma samples for these 50 patients. The earliest time point after 6 months of ART when a plasma specimen has greater than 400 copies/ml HIV-1 RNA will be defined as the time of virologic failure. We will compare the proportion of the 50 patients who have HIV-1 drug resistance mutations at the time of virologic failure and at the time of clinical/ immunologic failure. Specific Aim 2: To conduct surveillance for transmission of drug resistant HIV-1 strains in Port-au- Prince, Haiti. We will determine the number and type of HIV-1 genomic drug resistance mutations each year in 100 recently infected patients in Port-au-Prince, Haiti and follow trends over 4 years. HIV-1 strains isolated from recently infected individuals are representative of HIV-1 strains that are currently being transmitted within the general population. The proposed research will critically evaluate the WHO ART guidelines and provide drug resistance data to guide the choice of first-line and second-line ART regimens in resource-poor settings. During the conduct of this research, the applicant will be mentored by Drs. Jean Pape and Roy Gulick of Cornell University. He will also take intensive courses in clinical epidemiology and biostatistics.